Community ambulation for stroke survivors is a highly complex skill, requiring the ability to adapt not only to increased environmental complexity (relative to the clinic), but also to dual-tasking situations, such as walking while conversing. Indeed, the ability to walk in the community is a key determinant of community integration and participation. Very little is known about the impact of the environment on single-task and dual-task gait after stroke, even though it is well established that community-dwelling individuals with stroke experience profound cognitive-motor interference during dual-task walking and restricted community participation due to mobility limitations. The goals of the proposed research are to identify the differences between single and dual-task walking in the laboratory and single and dual-task walking in a real-world context (Aim 1), to determine whether the ability to flexibly prioritize attention during dual-task walking after stroke is impaired relative to matched healthy adults, and whether this ability is impacted by environmental setting (Aim 2), and to determine minimal clinically important effect sizes for dual-task interference effects in community-dwelling stroke survivors and matched healthy adults (Aim 3). The proposed research is an interdisciplinary collaboration involving experienced investigators in physical therapy, biomedical engineering, and psycholinguistics. Using a repeated measures design, gait and cognitive performance of 30 adults with stroke and 30 matched healthy adults will be assessed in single and three different variable-priority dual-task conditions (no-priority, gait-priority, cognitive-priority) in the laboratory and a real-world context (hospital lobby), and in two different dual-tak combinations (walking while talking, walking while performing an executive function task). We will use anchor-based and distribution-based approaches to estimate minimal clinically important difference (MCID) values of gait- related dual-task effects. This project will help us achieve our long-term goal to develop interventions to improve the locomotor adaptability and safety of community-dwelling stroke survivors. The proposed research will also enable us to identify the individuals most susceptible to clinically significant mobility disability in the realworld and acquire the data needed to adequately power a subsequent R01 intervention study targeting those most in need. This innovative study, concerned with stroke survivors' ability to walk in real-world environments, will yield crucial data currently unavailable in the literature.